Steve wrote:

This is just WRONG - read up about PROPER AC Bridges and not these toy constructions, and learn about resolution of parts per TRILLION, and accuracies of 20 parts per billion.(30 bit ACCURATE)
Who in their right mind would use a square wave to excite a bridge anyway ?

Steve,

is this your way to discuss? Just tear sentences from their context and offend people?

What I told is not at all 'just WRONG'! I wrote:

But even if we can handle these harmonics at input stage, ...

This included the methode of sinus bridge excitation.
I only focused on square wave excitation, because I wanted to compare simple DC excitation with another simple AC excitation methode, means square waves methode. This was already discussed in this thread, and I wanted to enlighten the consequences of this a bit.

Then I wrote:

If noise can exceed signal, lock-in methode is advantegous. But here again fast switching edges are coupled into inputs (or outputs) of linear amplifiers. Again drift, unlinearity and offset is introduced and it's very very difficult to obtain a dynamic range greater than 80dB, which represents a resolution of only 13bit!

And this corresponds to datasheet of AD630 from 'Analog Devices'. Here lock-in methode is described in detail. Lock-in methode indeed uses very fast synchronous switching from +signal to -signal from half wave to half wave. And this methode is used in special situations, where noise is so high, that signal seems to be burried in noise. AD630, which is rather expensive, achieves a dynamic range of 100dB, but only with careful calibration. With a standard design using discrete switches, due to charge injection only about 80dB is achievable. So, this is also right. I did not tell, that higher accuracies are impossible, but only that they are very very difficult to obtain with standard parts. And again, this performance data is referred to AC bridge excitation in combination with LOCK-IN METHODE.

May be, Steve, that I missed certain applications, where indeed higher dynamic ranges are possible. But this does not mean, that my text is 'just WRONG'. May be incomplete. And there's absolutely no reason for you to react such unfriendly!

By the way, if it is so easy to obtain 30bit accuracy with your AC bridge excitation, why are these circuits so seldom to find? How much do they cost? How much chips are needed to perform this? And are they really so competitive, that these so much simpler designs containing DC bridge excitation and chopper OPamps will disappear? This is what Erik wanted to know, isn't it?

Steve, it's no fun to contribute to this forum, when each reply is responded by you so aggressively! Ok, if your experience is so superior, that we can learn from you, then enter into a dialogue. But please in a friendly way...

What sense makes this forum, when we all the time offend another?

Kai